HAL Id: tel-01249379
https://tel.archives-ouvertes.fr/tel-01249379
Submitted on 4 Jan 2016HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
therapeutic hydrogels and films with controlled and
selective protein adsorption
Johanna Davila Ramos
To cite this version:
Johanna Davila Ramos. Syntheses and uses of modified polyelectrolytes for therapeutic hydrogels and films with controlled and selective protein adsorption. Other. Université de Strasbourg, 2012. English. �NNT : 2012STRAF005�. �tel-01249379�
ÉCOLE DOCTORALE DES SCIENCES CHIMIQUES
[ Institut Charles Sadron ]
THÈSE
présentée par
Johanna Davila
soutenue le 13 Avril 2012 pour obtenir le grade de
Docteur de l’Université de Strasbourg
Discipline/ Spécialité
:
Chimie-PhysiqueSyntheses and uses of modified polyelectrolytes for
therapeutic hydrogels and films with controlled and
selective protein adsorption
THÈSE dirigée par :
M. MESINI Philippe Directeur de recherche, Université de Strasbourg
MEMBRES DU JURY :
M. HOLL Yves Professeur, Université de Strasbourg et Président de
jury
M. DURAND Alain Professeur, Université de Lorraine, Nancy-LCPM
M. ROUCOULES Vincent Maître de Conférences, Université de Haute Alsace,
Acknowledgements
After all this year’s this thesis would not have been possible without the support of many people:
I would like to express my gratitude towards Philippe Mesini for giving me the opportunity to be a member in his research group, also for suggesting this interesting and challenging topic. Thanks to his valuable advices, encouragement and interest I was able to successfully complete my work.
I would like to extend my gratitude to Loïc Jierry for his scientific and personal advices, guidance, interesting and productive discussions during our meetings that helped me improve this work.
I would also like to thank to Alain Durand and Vicent Roucoules for having accepted to be part of the board of examiners and for providing interesting suggestions in my work. I am also grateful to Yves Holl for his participation as president of the advisory committee.
I would like to convoy my sincere thanks to Joseph Selb for the great help in rheological measurements analysis and the fruitful discussions for the interpretation of the results. Also thanks to Fouzia Boulmedais for working together and for all the help in the QCM measurements.
Many thanks to Francoise Hoegy and Armelle Chassepot who contributed to this work by completing the biological measurements. Thanks to Johan Longo, Camille Heid, Adrien Alibert and especially to Eric Gonthier for performing numerous experiments.
To my friends whom I had the fortune to meet, thanks for the discussions and the fun during lunch, coffee breaks and outside the lab. Without them it would have never been the same. I would like to especially thank to Andru the best labmate for always listening to me, Diana and Rebecca for their support and confidence and for teaching me a lot of things during all this time, Cesar for always making me laugh. I would like also to extend my warm gratitude to Patty who started this adventure with me a long time ago, thanks for all your moral support and for listening to me without complain, the next one will be you. “Vamos ya falta poco”
I would like to thank to my colleagues who accompanied me during all this time. Thanks to Rahul, Martin, Akkiz, Anna, Mirela, Tam, Christophe, Franck and to some of them that even far away supported me in someway Almeira, Maria, Gaby, Francine, Dominique, Edgar, Ivan for their huge support during the tough periods of my stay.
My deep thanks to my family. Without their constant support and advice I would not have been who I am today. Thanks for rekindling my dreams. Los amo.
With all my heart to my husband David who helped, supported, and always has faith in my abilities. Many thanks for all your love, patience and being beside me during my PhD work.
4 Table of Contents
ABBREVIATIONS 8
INTRODUCTION 11
1 CHAPTER 1: 15
MODIFICATION OF POLYELECTROLYTES TO THE BUILDUP OF FILMS FOR
BIOSENSOR AND MECHANORESPONSIVE SURFACES 15
1.1 Introduction of polyelectrolyte multilayer films (PEM) 16 1.1.1 Principle of the preparation of polyelectrolyte multilayer films 17 1.1.2 Mode of growth on polyelectrolyte multilayer films 19 1.1.3 Nature of the substrate 20
1.2 Multilayer films responsive to mechanical stimuli 20 1.3 Control of protein adsorption and cell-adhesion on surfaces 23
1.3.1 Strategy 23
1.3.2 Anti-fouling surfaces 24
1.4 Cell-adhesive surfaces 26
1.5 Biosensor systems using Polyelectrolyte multilayer films 28
1.5.1 Protein sensors 28
1.5.2 Immunosensors 29
1.6 Conclusion 32
2 CHAPTER 2: 34
POLYELECTROLYTE MULTILAYER FILMS DEVELOPED FOR BIOSENSOR
APPLICATIONS 34
2.1 Modifications of PAA by biotin and phosphorylcholine moieties 35 2.1.1 Synthesis of the EO linker 37 2.1.2 Coupling reactions of EO linker to Poly(acrylic acid) 39
2.2 Buildup of polyelectrolyte multilayer films by QCM 42 2.3 Serum and Streptavidin adsorption on functionalized films 44
5
2.3.2 PAA(EO)nBiotin ending films 45
2.3.3 PAA-PC-(EO)nBiotin ending films 48
2.4 Conclusion 50
3 CHAPTER 3: 52
CYTO-MECHANORESPONSIVE POLYELECTROLYTE MULTILAYER FILMS 52
3.1 Article 1: Cyto-mechanoresponsive Polyelectrolyte Multilayer Films 53 3.1.1 Article 1 (SUPPORTING INFORMATION) 59
4 CHAPTER 4: 82
ASSOCIATING POLYMERS FOR THERAPEUTIC HYDROGELS 82
Goals of the work 83
4.1 Relevance and suitability of synthetic polyelectrolytes for biomedical applications 84 4.2 Structure of associating polymers 87 4.2.1 Telechelic or end capped polymers and block copolymers 88 4.2.2 Multisticker copolymers 89
4.3 Structure and physical-chemical parameters that influence the thickening of modified
PAA and PMAA 91
4.3.1 Physical parameters 91
4.3.2 Structural parameters 95
4.4 Biological target: Pseudomonas aeruginosa and its lipase 97
4.5 Conclusion 99
5 CHAPTER 5: 102
SYNTHESIS AND USE OF ASSOCIATING POLYMERS FOR ENZYMATICALLY
RESPONSIVE GELS 102
5.1 Synthesis and characterization of the copolymers 104 5.1.1 Choice of the synthetic route 104 5.1.2 Synthesis of copolymers poly(tert-butyl methacrylate-co-n-alkyl methacrylate) 106
6
5.2.1 Solubility and gel formation 117 5.2.2 Rheological properties of the gels 118
5.3 Gel degradation 124
5.3.1 In vitro assays 124
5.3.2 In vivo assays 127
5.4 Conclusion 128
6 CHAPTER 6: 131
MATERIALS AND METHODS 131
6.1 Materials 132
6.1.1 Solvents and chemical reagents 132 6.1.2 Biomolecules- Enzymes, Proteins, Serum and Polyelectrolytes 133
6.1.3 Polyelectrolytes 133
6.2 Analytical methods and instruments 135 6.2.1 Chromatographic methods 135 6.2.2 Nuclear Magnetic Resonance (NMR) 135
6.2.3 FTIR spectrometer 135
6.2.4 Size exclusion chromatography (SEC): 135
6.2.5 MALDI-TOF MS 136
6.2.6 Elemental analysis 136
6.3 Gels formation 136
6.4 Rheology 136
6.5 Buildup of polyelectrolyte multilayer films 137 6.6 Quartz crystal microbalance (QCM) 137
6.7 Biological assays 138
6.7.1 In vitro assays 138
6.7.2 In vivo assays 138
6.8 Syntheses 139
6.8.1 Monomers for associating copolymers 139 6.8.2 Copolymerization reactions 140 6.8.3 Hydrolysis of tert-butyl groups 142
7
CONCLUSIONS AND PERSPECTIVES 151
ANNEXES 156 Annexe 1. 157 Rheology 157 Annexe 2. 161 Annexe 3. 163 Biomolecules 163 Annexe 4. 166 Rheological results 166 7 REFERENCES 169
8
Abbreviations
FTIR Fourier transforms infrared spectroscopy
GR grafting ratio
NMR nuclear magnetic resonance spectroscopy
PC phosphorylcholine or phosphorylcholine group
QCM quartz crystal microbalance
SEC size exclusion chromatography
AIBN azobisisobutyronitrile
Boc2O di-tert-butyl dicarbonate
CH2Cl2 dichloromethane
DMF N, N-dimethylformamide
EtOAc ethyl acetate
EDCI N-Ethyl-N'-(3-dimethylaminopropyl)carbodiimide hydrochloride
Et3N triethylamine
HCl hydrochloric acid
HOBt 1-hydroxybenzotriazole hydrate
Hz hertz
MgSO4 magnesium sulfate
NHS N-hydroxysuccinimide
NaN3 sodium azide
Ph3P triphenylphosphine
PAA poly(acrylic acid)
PAH poly(allyl amine hydrochloride)
PEI poly(ethyleneimine)
PSS poly(sodium-4-styrenesulfonate)
Tris tris(hydroxymethyl)aminomethane (buffer)
TsCl tosyl chloride
THF tetrahydrofuran
9
TLC thin layer chromathography
11
Introduction
12
During the past decades the development of new materials for biomedical applications has involved more and more polymers.1 This has been exemplified by the design of therapeutic macromolecules to increase the efficiency and the control in drug delivery. For instance, the efficiency of anticancer agents has been improved by using polymer-drug conjugates able to deliver them selectively in the targeted tissues.2-7 Polymeric systems have also allowed considerable progress in implants such as prostheses or temporary implants, especially those used for regenerative medicine and tissue engineering and involved polymeric systems have experienced a throbbing interest.8-10
Many of these materials can accomplish very precise and complex functions, thanks to a careful design and refined modification and functionalization. The functionality of these materials can be programmed, by changing their intrinsic properties and architecture, but cannot be modulated according external conditions. For instance a polymeric drug system can be designed to deliver insulin in a given duration, but it is more difficult to make the system deliver only in case of hyperglycemia. That’s why further work tends to obtain materials responsive to an external stimulus, for instance evolved polymers able to release drugs only on demands in a given stimulus. These stimuli can be simple physicochemical parameters such as temperature or pH. They can correspond to a biological situation: increase of temperature in case of fever, pH variation corresponding to the variation of pH along the digestive tract (acid in stomach, basic in duodenum) or to the acidic medium of lysosomes to release the drug only in the cell. More recent efforts try to elaborate materials able to respond to more sophisticated stimuli. For example recently magnetic field sensitive carriers have been developed to deliver drugs on demand or on a given spot of the organism.11, 12
The doctorate research presented here contributes to the elaboration of stimuli responsive materials based on modified polyelectrolytes.
In the first chapter we provide background information about the major elements required for the modification of surfaces with polyelectrolyte multilayer films.
In the second chapter, we have developed polyelectrolyte multilayers to design surfaces able to bind a given protein very selectively and repelling all other. This property is sought to form biosensors. One has also taken advantage to design surfaces that respond to mechanical
13
stretch. The expected response is to trigger cell adhesion under stretch, while under rest the same surface is anti-adhesive. The results of these studies are presented and discussed in chapter 3.
Subsequently we have presented a detailed study of the structure of associating polymers and the changes of their properties under the influence of different parameters. We have mentioned the applications as drug delivery systems that result from their properties.
The chapter 5 describes our efforts to obtain a polymeric gel with a rheological response when exposed to pathogen bacteria. The stimulus is constituted by microbial exoproteins that signal the presence of bacteria. The sought response is the viscosity decrease and resolubilization of the gel. Such polymer could be used as a drug carrier able to store high antibiotic doses and release them only in the case of infections.
15
1
CHAPTER 1:
Modification of polyelectrolytes to the
buildup of films for biosensor and
16
First this study is related to the modification of polyelectrolyte to be used in the buildup of films. During my work we prepared poly(acrylic acid) PAA with two bioactive molecules, RGD and biotin. This functionalizations endows the films, incorporating the modified polymer with new properties necessary to form:
- Biosensor surfaces
- Mechanoresponsives multilayer films
This bibliographical part will first describe the key elements for the buildup of polyelectrolyte multilayer films (PEM) will be briefly described in this chapter. Then the applications of PEM will be gathered dedicated to biosensing surfaces and mechanoresponsive materials.
1.1 Introduction of polyelectrolyte multilayer films (PEM)
The multilayer systems were introduced in the 1930s by Irving Langmuir and Katherine Blodgett describing nanostructured films called “Langmuir-Blodgett” (LB) films.13, 14 This technique consists in the construction of a monolayer of amphiphilic molecules at the air-water interface. This film is then transferred onto a solid support. The process can be repeated several times to obtain multilayer films. Yet the LB technique presents several limitations and inconveniencies, the films are not stable, they cannot be formed on substrates and have limited sizes. In the 1980s new systems were developed as an alternative to LB films, by self-assembled monolayers (SAMs) on silane-SiO2 or by covalent15 and coordinate16 chemistry. However, these
systems cannot produce high-quality multilayer films.
The development of nanostructured films based on electrostatic interactions was carried out by Iller.17 The construction of multilayer films required colloida anionic and cationic particles, but this method was never proved. This technique relies on the electrostatic attraction between oppositely charged molecules seemed to be a good candidate with good driving forces for multilayer build-up. At the end of the 1980s this concept was applied in some studies on adsorption of proteins18 and metallic’s colloids.19
It was in the 1990s that Decher20 demonstrated the possibility to build up films by the alternating physisorption of oppositely charged polyelectrolytes, so called layer-by-layer method
17
(LbL). This kind of construction allows the deposition of many organic and inorganic substances and organic as DNA21, on any substrate from gold to silicon, through a simple stepwise process. Later the LbL was extended to other interactions than simple electrostatic, and some films have been successfully made via hydrophobic,22 hydrogen bonds23 or covalent bonds.24
Here we review only the PEM built from electrostatic interactions. We focus on the understanding of the properties of PEMs necessary to guide the functionalization of the polyelectrolytes with bioactive molecules.
1.1.1 Principle of the preparation of polyelectrolyte multilayer films
The technique is based on the electrostatic interactions between cationic and anionic species and the entropic release of coutenrions. The substrate is brought into contact with dipped in a solution of polyion, followed by a rinsing step to eliminate the excess or the weakly adhering polyelectrolyte chains.20, 25, 26 Then the substrate is brought in contact with a solution of oppositely charged polyelectrolytes and rinsed again. A representation of the buildup of a multilayer film is displayed in Figure 1. 1. These steps are then repeated until the desired film thickness is obtained.
Figure 1. 1. Principle of the construction of polyelectrolyte multilayer films by dipping. Steps 1 and 3 represent the adsorption of polyanion and polycation, respectively. Followed by rising steps 2 and 4.20
18
The PEM can be obtained by different deposition techniques:
The dipping method is represented in Figure 1. 1. This is historically the first method and it is still the most widely used. The deposition requires between 5 and 20 minutes for each polycation/polyanion deposition cycle including the rinsing step. 20, 27 This technique can be coupled to measurements by quartz crystal microbalance (QCM) and optical waveguide lightmode spectroscopy (OWLS), by replacing the dipping steps by letting the different solutions flow through to measurement cells.
The spraying method consists in spraying the solutions horizontally on the surface held
vertically. This technique considerably speeds up the process.28 Porcel et al.29 demonstrated that it is possible to produce films by simultaneously spraying both polyelectrolytes which reduces even more the time of film construction. The layer thicknesses obtained by spray Layer by Layer (LbL) are usually less than those obtained by dipping.30, 31(Figure 1. 12(a))
Spin coating can also be used to build up PEM.32, 33 It can produce more homogenous films, Figure 1. 12(b).
Figure 1. 2. Schematics representation of LbL deposition techniques on substrates. a) spray LbL. b) spin coating. The process is repeating several times until desired layer number is obtained.
When constructing PEM’s by the LbL method both the mass and thickness of the film increase with the number of deposition steps. There are two types of film growth, linear and exponential, which will be reviewed in the following section.
19
1.1.2 Mode of growth on polyelectrolyte multilayer films
The two types of growth regimes are represented in Figure 1.3, where the variation of thickness with the number of the deposited layers is shown.
Figure 1. 3. Evolution of the polyelectrolyte thickness. Linear versus exponential growth. ( ) PEI(PGA/PLL)n and ( ) PEI(PSS/PAH)n as a function of the numbers of added layer pairs.34
In the linear growth regime, the thickness increment of the multilayers is constant with the number of alternative steps of deposition of polycation and polyanion. In this case, polyelectrolytes from the solution interact exclusively with the outer layer of the film without diffusing into the architecture. This is observed for the poly(styrene sulfonate (PSS)) and
poly(allylamine hydrochloride (PAH)) (PSS/PAH)34, for (PAH/PAA)35 or
poly(diallyldimethylammonium (PDADMA)) (PDADMA/PSS)36 films.
If the film thickness and the amount of adsorbed polyelectrolytes increase with the number of deposition steps, the film shows an exponential growth. This behavior is displayed for example by (PGA/PLL)34 and (PLL/HA).34, 37 In those PEM’s, part of the polyelectrolyte can diffuse freely throughout the thickness of the film. So the proposed model for the exponential growth is as follows: an amount of polyelectrolyte is adsorbed on the top of the film. This amount is limited by the charge compensation of the previous layer. But an additional amount migrates inside the film and freely diffuse throughout the thickness of the film. In the next step, when the oppositely charged polymer is brought into contact of the film, the free previous polyelectrolyte migrates back to the top and increases the amount of polymers that can be
20
compensated. This leads to a thickness increment proportional to the thickness of the multilayer and hence to an exponential growth.
1.1.3 Nature of the substrate
There are many substrates for the construction of polyelectrolyte multilayers. The surfaces can be flat or spherical. In this thesis we have worked with flat surfaces. Some types of surfaces are mineral (glass slide20, quartz38), metal (sintered titanium39), silicon40 and organic (polyethylene41).
In this work polydimethylsiloxane (PDMS) has been used as substrate in order to stretch the. This polymer is of particular interest because of its elastic properties, low toxicity, chemical inertness and biocompatibility. This property have been exploited in the domain of medical implants., especially in vascular grafts and breast implants.42 Moreover it has been used to control and enhance cell adhesion.43, 44
It is not easy to build PEMs on PDMS: the surface is not charged. So the adsorption of the first layer of polyelectrolyte is quite delicate and relies exclusively on Van der Waals interactions. The adhesion to the silicon depends on the choice of polyelectrolytes. The films based on (PSS/PAH) are easily deposited on silicon surfaces45, while (PAA/PAH) films shows a high surface roughness.46
1.2 Multilayer films responsive to mechanical stimuli
During the last years much research has been dedicated to materials with responsive surfaces. The design of materials that are sensitive to external stimuli such as pH47, temperature48, and light49 has been widely investigated. However, less attention has been given to films that respond to mechanical stimuli.50
Genzer and coworkers51 have stretched silicone sheets, modified them by attaching perfluoroalkane chains and relaxed them after modification. In this way, they were able to develop “mechanically assembled monolayers” which significantly decrease water contact angle upon relaxation. The final hydrophobicity strongly depends on the degree of stretching before chemical treatment.
21
One of the first examples of adaptative multilayer films has been reported by the group of P. Schaaf.52 Nafion (Naf), a hydrophobic polyelectrolyte, was used to build the following film on PDMS substrates: (PAH/Naf)4-(PAH/PAA)2. At rest the film has a water contact angle of 45°.
When the film is stretched to 220 % the contact angle increases to 100°. This property is attributed to the exposition of hydrophobic Naf layers at the surface when the film is stretched. (Figure 1.4)
Figure 1. 4. left: schematic representation of the film that exhibits a reversible hydrophobic/hydrophilic transition upon stretching; right: evolution of the contact angle of PEI(Naf/PAH)4Naf(PAH/PAA)2 multilayer. The angle variations are observed for films ending with a
hydrophilic layer (PAH/PAA) (A) deposited on top of a hydrophobic one whereas no changes in the contact angle are observed when the hydrophobic film (PAH/Naf) (B) is deposited on top of hydrophilic one.52
This system proved to be reversible and reproducible over successive stretching and relaxation cycles. The authors have hence achieved a system that responds to a mechanical stimulus by reversibly switching from hydrophilic to hydrophobic.
Another mechanosensitive system was recently reported by Möhwald and coworkers.53 It is based on polyelectrolyte multilayer films from polystyrene sulfonate modified with pyrene (PSS-PY) and poly(diallyldimethylammonium)chloride (PDADMA) on PDMS. When the film is stretched, pyrene fluorescence at 457 nm decreases. This fluorescence corresponds to the emission of the excimer. When the film is stretched, the ratio of associated pyrenes diminishes because the polymer chain are forced into an uncoiled state. (Figure 1.5)
22
Figure 1. 5. Schematic representation of the effect at unstretched and when stretched. (PSS-PY) with x = 1 and 3%.(Green) PSS-PY polyelectrolyte (Red) PDADMA molecules.
The transformation of mechanical stimulus such as stretching into a chemical or biological reaction is a process that is widely used by nature and that is called transduction.54 When subjected to mechanical forces, some proteins change their conformation. This change allows the exhibition of hidden active sites, called cryptic sites, that become accessible. For instance fibronectin is able to expose such cryptic sites under stretching.54-56 Our work has sought to develop films mimicking this property, for instance films containing active compounds which are inaccessible in the non constrained state and become accessible in a reversible way when the surface is stretched.
More recently Mertz56 et al. describes films with biocatalytic activity induced by stretching. The multilayer architectures were built up on a silicon sheet, Figure 1.6. The films consist in a first exponentially growing multilayer (PLL/HA)15/PLL/ALP/(PLL/HA)15 and
covered by (PDADMA/PSS)6 that acts as a barrier. ALP (alkaline phosphatase) was embedded in
the films. This enzyme hydrolyzes FDP (fluorescein diphosphate). This interaction (ALP-FDP) was displayed when the films is stretched leading to green fluorescence emission. This fluorescence suggests that the cryptic sites have become accessible upon stretching. At rest, only a weak fluorescence was observed.
23
Figure 1. 6. Scheme of the mechanically sensitive biocatalytic coating. (i) At rest the catalysis is off. (ii) under stretching the enzymes are exposed, the biocatalysis is active leading to formation of fluorescein. (iii) Inhibition of enzymes due to strong increase of the concentration of fluorescein and phosphate ions. (iv) Rinsing step to reactivate the enzymes. When the system is brought back to its initial state (i) the enzymes are again masked and the catalysis is desactivated.56
These types of architectures demonstrate the utility of PEM to mimic biological systems. Until now, very few studies have been carried out with the functionalization of polyelectrolytes with different molecules for specific biological applications.
1.3 Control of protein adsorption and cell-adhesion on surfaces
1.3.1 Strategy
PEM can coat “virtually” any surface. An interesting perspective is to use them to coat medical implants. One of prime importance for all applications of implants is to control their interactions with the biological environment. For those applications one seeks two main properties.
- It must be protein repellant: this property is necessary to render the implant blood compatible for instance in catheters or heart valves to avoid thrombosis or bacterial infections.
- It must be adherent for the cells: this property is necessary in tissue repair or in bypass systems.
24
Some applications require both properties (anti-fouling properties and cell adhesion) simultaneously. The PEMs are very prone to adsorb proteins. The proteins can be adsorbed on any polyelectrolytes by electrostatic interactions, depending on the charge of the protein. However even in the case of unfavourable charges, the protein can absorb. Ladam et al.57 describe the adsorption of negative serum albumin (HSA) that is negative onto positively or negatively charged polyelectrolyte film. On the positive film HAS adsorbs and the thickness of the adsorbed protein layer exceeding several times the typical protein size. On the negative films, HSA still adsorbs and in this case only one monolayer is adsorbed. This behavior is due to hydrogen-bonding and hydrophobic interactions.58
Besides physicochemical adsorption, some proteins can be bound to PEM´s by ligand-protein interaction. This can be done by incorporating ligands in the film, which they are accessible to one protein exclusively, as biotin-streptavidin.59,60 Our strategy is the design of surfaces that can be specific for protein adsorption and adherent to cells.
The next section presents some of antifouling films.
1.3.2 Anti-fouling surfaces
In this chapter we present two antifouling polymeric coatings: the first is based on PEO, the second one on phosphorylcholine (PC) groups.
Surfaces coated or grafted with PEO were introduced in the 80s. For high molecular weights and high grafting densities, PEO forms a brush conformation that sterically impedes protein adsorption.61 The reduction of protein adsorption is related to the grafting density of PEO, to their surface density and to their length.
In order to coat surfaces with PEO, block copolymers with a hydrophobic block are used to absorb the substrate on hydrophobic surfaces. Adhesion to hydrophobic substrates can also be obtained using methacrylate backbones with alkyl side chains.62 Polyelectrolytes can be easily modified to form antifouling PEMs: brush copolymers of PLL modified with 5 mol % of PEO (PLL-g-PEO) on metal oxide surfaces effectively reduce the adsorption of blood serum and human fibrinogen.63, 64 Boulmedais et al.65 have studied multilayers based on (PLL/PGA) and
25
observed a strong reduction of protein adsorption with a hydrophilic multilayer of (PLL/PGA-g-PEO).
The other group of antifouling species, less studied than PEO, are compounds containing phosphorylcholine (PC). Many phospholipids compose the cell membranes and can have either uncharged, positive or negative head groups. But it has been noticed that in blood cells, PC is the major membrane constituent of the external membranes. PC groups lead to blood compatibility because they reduce the adsorption of proteins. 66, 6, 67 Later it has been found that many other zwitterionic groups induce protein resistance, e.g. sulfobetains having equal amounts of N(CH3)3+ and SO3- groups.
In our group, Reisch et al.68, 69 have modified many polyelectrolytes with side chains ended by PC (Scheme 1.1). Their inclusion in PEMs strongly reduces protein adsorption even under stretching.
Scheme 1. 1. Chemical structure of PAA(EO)3PC.
Films of PEI(PSS/PAH)5/PAA(EO)3-PC where PAA(EO)3-PC is the modified PAA
depicted Scheme 1.1 proved to be fully protein repellent for rates of modification of 25 %.69 The same film was built on PDMS and its antifouling capacity was measured under rest and after stretching. The systems was stretched to 100 and 150% after exposure to fluorescein labeled albumin (AlbFITC) and the amount of adsorbed protein was measured by fluorescence microscopy. The results are illustrated Figure 1. 7. The fluorescence slightly increases when it is stretched. One has to deposit two PAA(EO)3PC top-layers to eliminate all adsorption under
stretching. The antifouling properties were also tested with PAA-(EO)3 with a modification
26
Figure 1. 7. Fluorescence micrograms of protein adsorption of PEI/PAH/PSS/PAH, PEI(PSS/PAH)5
(PAA-(EO)3PC)n n = 1 or 2 Treated onto silicon surface after adsorption of AlbFITC and rinsing.
These types of systems are one of the main trends in biomedical applications involving modification of polyelectrolytes with bioactive molecules. They are e.g. used as thin films for immunoassays, chemical and biological sensors70, 71, drug screening, tissue engineering, films to immobilize DNA, proteins and enzymes for biosensors applications. 72
1.4 Cell-adhesive surfaces
When cell adheres to a substrate, it involves complex biochemical mechanisms. Especially, in the living tissues the cell adhere to the extracellular matrix via membrane proteins called integrins.73 These proteins mediate cell adhesion by binding to different proteins of the extracellular matrix (ECM).
The tripeptide RGD (arginine-glycine-aspartic acid) is a ligand of integrins and can stimulate cell adhesions when it is adsorbed on surfaces. This sequence was identified in many other ECM proteins e.g. in fitronectin, fibrinogen, collagen, laminin, etc, that trigger cell adhesion.
27 Figure 1. 8. Adhesion and proliferation of the cell.
Therefore, functionalization of polyelectrolyte multilayers with RGD endows them with cell adhesion capacity . This peptide can be easily linked to polymers via covalent amide bonds. The PAH proved to be excellent for binding RGD.74 When RGD was covalently immobilized in films of alkyne or azide groups (PEG-Alk/PEG-Az)5 it promoted the adhesion and proliferation
of cells.75
In similar systems RGD is grafted to (PLL-g-PEG/PEG-RGDx%) with (x% = 1, 4, 11, 58) The structure is represented in Scheme 1.2 and PEMs with this modified polymer on top were built on a Ni2O5 surface. The construction of the film and cell adhesion was monitored by the
OWLS apparatus. 76 This film promotes the adhesion of human dermal fibroblasts. In these examples, the films are functionalized at the same time with PEG in order to repel proteins. This double functionalization affords a film that is antifouling for most of the species, but adherent for the targeted cells.
28
1.5 Biosensor systems using Polyelectrolyte multilayer films
A biosensor has three major components: the bioreceptor that is a biomolecule that recognizes a given target, the target molecule or sensor molecule, and the transducer which converts the recognition into a measurable signal as (optical, piezoelectric, electrochemical, etc). The sensor molecule can consist of a large variety of molecules, e.g. enzymes, antibody, etc.
The principle of biosensors was introduced in 1962 by Leland et al.77 . His system was able to measureme blood glycemiae and described the first enzyme electrode with immobilized glucose oxidase.
LbL is readily used to prepare biosensors. Multilayer thin films including oxidoreductases (glucose oxidase78, lactate oxidase79,) or other proteins80 by the alternating deposition of proteins and oppositely charged polymers have been prepared. These protein-containing LbL films can be immobilized on electrode surfaces. When the reaction takes place between the enzyme and its substrate,. the oxidation is converted in electric signal via the electrode. This technique has proven to be relatively simple and versatile.
1.5.1 Protein sensors
Many PEM biosensors have been tested with biotin as the sensor and avidin as the bioreceptor. Indeed, biotin-strepatvidin presents the strongest non-covalent biological interaction known, with binding constant of 1015 M-1. With this strong interaction, it has been easy for many teams to develop films able to bind biotin or avidin.57, 81-83. Biotin is an acid and can be readily coupled to polyelectrolytes, as illustrated in Scheme 1.3.
29
Roucoules and coworkers.84, designed surface responsive materials based on PDMS substrates. PDMS was functionalized with carboxylic acid groups by a pulse plasma treatment in the presence of maleic anhydride. COOH groups were used to covalently attach PEG (MW = 2000) and PEG-biotin groups. The studied the non-specific adsorption of proteins, under stretch and at rest in the presence of avidin. Figure 1. 9 illustrates the behavior of the PDMS surface.
Figure 1. 9. Schematic representation of the PEG-Biotin covalently attached on the PDMS surface in contact with avidin. On the right representation of the surface at rest and on the left the surface under stretch.
The studies of the surface demonstrate that the surfaces are anti-fouling when they are not stretched. This property varies with the elongation of the substrate. When the substrate is stretched at 100% non-specific adsorption of proteins takes place, since the surface is now accessible. Nevertheless, these types of surfaces are precursors for the design of mechanically responsive materials able to resist non-specific adsorption at rest and under stretch.
In one hand PEM can be modified to inhibit the non-specific adsorption of proteins proteins interactions with the surface. On the other hand they can be modified to bind specifically given proteins such streptavidin or avidin. This double capacity offers an interesting perspective to form biosensors.
1.5.2 Immunosensors
Immunosensors exploit the molecular recognition of antigens by antibodies to form a stable complex. The antibodies constitute a class of proteins called immunoglobulins, which are produced by B-cells and they can bind to a foreign substance/molecule (antigen). Immunoglobulins are composed of four polypeptide chains that are connected by disulfide bonds and noncovalent interactions, yielding a Y-shaped protein. There are five types of antibodies called immunoglobulin (Ig), IgM, IgA, IgD, IgE and IgG. They are used in a wide range of
30
applications e.g., medical implants, drug delivery, clinical diagnostics. Their major application is the detection of drugs in blood or urine. For this purpose, the antibodies are labeled with an enzyme, a fluorescent compound or biotin providing a detectable complex able to be quantified. The construction of these systems with a biological recognition layer is fundamental for their use as biosensors.
The first PEM immunosensors used monoclonal antibodies anti-immunoglobulin G(anti-IgG) embedded in LbL films of (PSS/PAH)2 for the immobilization of the antibodies.85 The
assembly process and the binding capacity of anti-IgG-IgG were monitored using quartz crystal microbalance (QCM) and surface Plasmon resonance (SPR). Other approaches have been reported using LbL assembled on colloidal gold particles.86
The performance of immunosensors must be improved by lowering the nonspecific adsorption. As described above, it can be achievd with PEO. Recently,87 a biosensor with biotinylated PEO grafted to poly(L-lysine), (PLL-g-PEG/PEG-biotin+PLL-g-PEO) provides a
surface that is highly resistant to nonspecific adsorption from serum. The film construction was studied by optical waveguide light mode spectroscopy (OWLS). The copolymers of (PLL-g-PEO/PEO-biotin+PLL-g-PEO) are assembled on the negative surface of niobium oxide (Nb2O5).
31
Figure 1. 10. (a) Schematic representation of the different steps in the model of immunoassay on PLL-g-PEO/PEO-biotin surface. 1) Surface resist to nonspecific protein adsorption. 2) specific adsorption of streptavidin. 3) αRIgG-biotin binds to streptavidin (antibody). 4) RIgG(antigen) binds αRIgG-biotin. (b) Sequence of streptavidin, antibody and antigen adsorption on PLL-g-PEG/PEG-biotin surface by OWLS measurement.87
The biotinylated (PLL-g-PEG/PEG-biotin) is taken as a model for immunoassays. The presence of streptavidin on the surface of the multilayer film leads the interaction biotin-streptavidin. Once the streptavidin is immobilized on the film, it binds to the antibody (biotinylated goat rabbit immunoglobulin (αRIgG-biotin)), which in turn interacts with the antigen (RIgG). Figure 1. 10
Figure 1. 11(b) shows that the adsorption of αRIgG-biotin is lower than that of streptavidin. This behavior can probably be explained by steric effects of the more closely packed layers of PEG-biotin, blocking biotin or streptavidin free sites. Another reason may be the flexibility of the PEG-biotin chains, allowing the ligands to bind the four sites of the streptavidin, blocking the sites to the adsorption of the antibody. This behavior is represented in Figure 1.11.
32
Figure 1. 11. Shematic representation of possible streptavidin configurations. a) PEG-biotin binding to a single streptavidin in there four free sites. b) Steric effects making less accessible the streptavidin or biotin.
To conclude the LbL technique can readily form films with very elaborate surface properties.
1.6 Conclusion
To summarize polyelectrolyte multilayer films can be designed to become anti-fouling, biosensing or cell adhesive. The best known groups that inhibit the non specific adsorption of proteins or cell adhesion are poly(ethylene glycol) and phosphorylcholine.
In addition, PEM can be designed to change their properties under a mechanical stimulus. In the present work, we have elaborated new multilayer films with new biological properties. The first films are model biosensors, that can bind strongly streptavidin, but to repel all other proteins. This study will be described in chapter 2.
The second are mechanoresponsive films for cell adhesion. They embed hidden RGD moieties that become exposed to the surface under stretching. This part of the project will be presented in the chapter 3.
34
2
CHAPTER 2:
Polyelectrolyte multilayer films developed for
biosensor applications
35
This chapter describes the design and the buildup of polyelectrolyte multilayer films (PEM) able to bind selectively one protein and in the same time able to repel all others.
We have chosen streptavidin as a model of the protein specifically bounded. As discussed in the previous chapter, biotin can be tethered on PEMs and can bind the protein through a strong interaction. We have developed PEM films with a top layer composed of poly(acrylic acid) (PAA) grafted with biotin. This ligand has been covalently attached to PAA through an oligo(ethylene oxide) (EO)n spacer having different lengths (3, 9 and 18 EO units) or no spacer
for comparison sake. (EO)nBiotin was attached with different grafting ratios (GR), roughly 1%,
5%, 10% and 25%.
Furthermore, we modified PAA both with biotin and phosphorylcholine (PC) groups, to compare their response with streptavidin and their resistance to the non specific adsorption of other proteins, e.g fetal bovine serum (serum). Recently68, 69, our group proved that modified surfaces with PEM having PC groups on the surface can strongly reduce the adsorption of serum. These results were obtained with PAA functionalized with 25 % of phosphorylcholine groups (PAA-PC).
The first part of this chapter is dedicated to the preparation and characterization of modified PAA. The second part concerns the use of modified PAA as a last layer of PEM films to create a potential biosensor.
2.1 Modifications of PAA by biotin and phosphorylcholine moieties
In order to study anti-fouling property of films and the specific adsorption of streptavidin, we prepared two kinds of modified PAA:
i) PAA modified with EO groups bearing biotin moiety (PAA(EO)nBiotin), with different
GR roughly 1 to 25 %.
ii) PAA modified with PC and (EO)n groups bearing biotin (PAA-PC-(EO)nBiotin) with a
GR in PC at ≈ 25 % and in biotin at ≈ 1 %.
In both strategies the length of (EO)n was varied from n = 0 to n = 18. The selection of
36
The chemical structure of the modified polyelectrolytes to PAA(EO)nBiotin and
PAA-PC-(EO)nBiotin is given in Scheme 2.1.
37 General synthetic pathway is
a) Synthesis of PC and Biotin linkers
b) Coupling of PC and biotin linkers to PAA
Strategy 1.
Strategy 2.
2.1.1 Synthesis of the EO linker
To functionalize PAA, EOn linkers were prepared with a PC or biotin group on one side
and a protected amine on the other side. Biotin linkers synthesis is described for each length of EOn in Scheme 2.2 and 2.3.
38
The synthesis of BocNH(EO)3Biotin was done in 6 steps (Scheme 2.2). The linker is
acylated by biotin on one side and the other side is reserved to react with carboxylic acid of PAA in the last step. It starts from commercially available tetraethylene glycol. Alcohol functions were replaced by tosyl groups to afford the bis(tosyl) 1.89 Tosyl groups as leaving groups have been succesfully used in the modification of oligo and poly(ethylene glycol).90 This reaction yields to bi-substituted product with up to 40%.
Scheme 2. 2. Synthesis of BocNH(EO)3Biotin.
Bis(azide) 2 was obtained by the reaction with sodium azide in DMF with a yield of 88% from 1, according to established methods.89, 91 The mono-reduction to amines was achieved with one equivalent of triphenylphosphine (PPh3) (Staudinger reduction).89, 92 This reaction is
conducted with EtOAc/acidic water. This set-up allows the solubilisation of the monamine in water and prevents its to bireduction. It affords 3 in 81% yields. The obtained amine group was then protected with (Boc)2O to obtain 4. The azide group, on the other side was reduced by using
PPh3 to afford 5.93 The last step was acylation of the free amine by NHS-biotin in DMF to obtain 6.94, 95
The synthesis of BocNH(EO)9Biotin 8 and BocNH(EO)18Biotin 9 were done in one step.
Indeed, the starting compounds BocNH(EO)9NH2and NH2(EO)18NH2 are commercially available
from Aldrich. The synthesis of 9 was done from the bis(amine) with one stoichiometric equivalent of (Boc)2O and one equivalent of NHS-biotin. The yield is 40 %.The reactions are
39
Scheme 2. 3. Synthesis of BocNH(EO)9Biotin 7 and BocNH(EO)18Biotin 8.
The PC linkers were synthesized according to the procedure of Brockerhoff et al.96, 97 (Scheme 2.4). The synthesis consists in two successive one-pot reactions: reaction of POCl3 with
BocNH(EO)3H (in the presence of NEt3), followed by the reaction of the formed compound with
with choline tosylate (in the presence of pyridine). After hydrolysis, 9 was obtained with 50 % of yield.
Scheme 2. 4. Synthesis of BocNH(EO)3PC 9.
2.1.2 Coupling reactions of EO linker to Poly(acrylic acid)
EO linkers were covalently coupled to PAA (Mw = 100 000 from Sigma Aldrich) with the coupling procedure described by Reisch et al.97 The GR is defined as a percentage of modified carboxylate functions of PAA. The linkers, described below, were quantitatively deprotected by acidic treatment (HCl, 2.5 M) and coupled in situ with PAA in water at pH 5, with an excess of EDCI and HOBt. The polymer was purified by dialysis against water during 5 days.
40
a) PAA(EO)nBiotin
We synthesized PAA(EO)nBiotin with different lengths of ethylene oxide (n = 0, 3, 9 and
18) and different percentages of modification GR (1, 5, 10 and 25 %). The reactions are illustrated in Scheme 2.5 for EOn with n = 3, 9 and 18. For n = 0, the modification of PAA was
achieved with biotin-hydrazide. Scheme 2.6
Scheme 2. 5. Coupling reaction between PAA and BocNH(EO)nBiotin, (n = 3, 9, 18) and GR (1 , 5, 10
and 25 %).
Scheme 2. 6. Coupling reaction between PAA and Biotin hydrazide, GR ( 1, 5 and 10 %).
The polymers were obtained with an average yield 80-90 % (Table 2.1). Since all the polymers are water soluble, the purification can be obtained by dialysis and gave satisfactory purities.
b) PAA-PC
We first followed the procedure illustrated in Scheme 2.7 described by Reisch et al.97 to modify PAA with BocNH(EO)3PC 9 with a GR of 25 %. As it was described above the linker
was first deprotected under acidic conditions and coupled to PAA. The PAA-PC was obtained with a yield of 80 % after dialysis.
41
Scheme 2. 7. Coupling reaction between PAA and BocNH(EO)3PC to obtain PAA-PC 25 %.
BocNH(EO)nBiotin (n = 3, 9, 18) was grafted at 1 % on PAA-PC, by the same method as
above to afford PAA-PC-(EO)nBiotin , 25 %: 1 % (Scheme 2.8). PAA-PC-Biotin was
synthesized by coupling reaction between biotin-hydrazide and PAA-PC.
Scheme 2. 8. Coupling reaction between PAA-PC 25 % and BocNH(EO)nBiotin 1 %.
2.2.3 Characterization of modified PAA
GR of PAA-PC polymers was determined by 1H NMR from the integral ratio of the signals at δ = 4.23 ppm of PC (-CH2OP-) and at δ = 230 ppm of PAA(–CH-). The biotin content was
determined by the integral of signals at δ = 2.30 ppm of PAA (–CH-) and the signals at δ = 4.62 - 4.43 ppm of biotin (S-CH2). The spectra obtained for PAA modified by biotin are consistent with
the ones found in the literature for products bearing EO and biotin moieties.94 The elemental analysis could also provide a measurement of the biotin GR by the sulfur mass, since biotin
42
contains one atom of sulfur (see materials and methods). These measurements were consistent with those from NMR.
2.2 Buildup of polyelectrolyte multilayer films by QCM
We first followed the adsorption of functionalized PAA by quartz crystal microbalance (QCM). To avoid the influence of the substrate, we first built a PEI(PSS/PAH)3/PAH/PAA as a
precursor film. On top of which the modified PAA to be studied was adsorbed. The adsorptions of PEI, PSS and PAH are quite fast, about 5 min. There were, followed by rinsing steps of 5 min. The adsorption and the rinsing step of modified PAA were performed until the stabilization of the QCM signal.
As an example, the buildup of PEI/(PSS/PAH)3/PAA/PAH/PAA-PC-(EO)nBiotin film is
given in Figure 2.1. The evolution of the frequency shift ∆f at 15 MHz (∆f3/3) is representative of
the film buildup and is associated to a mass increase at each deposition steps. The measurement of all the frequencies allows to apply the Sauerbrey relation to calculate the mass and the thickness of the film.
Figure 2. 1. Evolution of the normalized frequency shift at 15 MHz (third harmonic) during the buildup of PEI(PSS/PAH)3/PAA/PAH/PAA-PC-(EO)nBiotin film as a function of time.
0 20 40 60 80 100 120 140 160 -500 -400 -300 -200 -100 0 ∆ f3 /3 (H z ) time (min) addition of PAA-PC/(EO)3Biotin
rinsing PAA PSS PSS PSS PAH PAH PAH PAH
43
The increment of thickness obtained for the different modified PAA and PAA is reported in Table 2.1.The thickness of modified PAA increases with the increase of GR and the length of EOn linker. This can be explained by the deposition of bound water on the film, due to the strong
hydration of ethylene glycol groups.69
Table 2. 1. Thicknesses of modified PAA adsorbed on PEI(PSS/PAH)3PAA/PAHcalculated from QCM
data by applying the Sauerbrey relation. GR in % GR in %(1H NMR) Yield % Thickness nm PAA 0 - - 7.7±0.7 PAA-PC 25 24±2 80 24±0.2 PAA-B 1 1±2 83 11.5±0.9 5 6±4 86 12.8±0.2 10 12±4 89 16±1.1 PAA(EO)3Biotin 1 1±2 86 8.6±0.9 5 4±2 93 8.5±0.7 10 11±4 87 14.4±0.4 25 25±5 89 18.4±0.6 PAA(EO)9Biotin 1 1±2 85 8.6±0 5 5±2 93 13.3±1.2 10 12±4 91 17.4±0.4 25 24±5 87 31.7±1.3 PAA(EO)18Biotin 1 1±2 93 9.7±0.2 5 6±2 89 16.5±2.4 10 9±2 81 21.2±1.1 25 24±2 83 61.7±1.1 PAA-PC/Biotin 25/1 24/1±2 94 25.7±0.8 PAA-PC/(EO)3Biotin 25/1 24/1±2 96 27.3±6.6 PAA-PC/(EO)9Biotin 25/1 24/1±2 95 28.3±0.3 PAA-PC/(EO)18Biotin 25/1 24/1±2 96 28.8±1.1
44
2.3 Serum and Streptavidin adsorption on functionalized films
Once the modified PAA adsorbed on the top of the films, we measured the amount of streptavidin and serum adsorption.
2.3.1 PAA ending films
As a blank, we determined the adsorption of serum and streptavidin on PEI(PSS/PAH)3PAA/PAH/PAA. On these films, the adsorption is necessarily non-specific since
it contains no biotin.
The QCM traces during adsorption of serum and streptavidin are shown in Figure 2.2, respectively. A large amount of serum (1252 ng/cm2) is adsorbed on PAA ending films (Figure2.2(b)) On the contrary, no adsorption of streptavidin is observed, Figure 2.2(a). Thus, PAA adsorb non-specifically serum proteins. It is expected that streptavidin does not bind specifically to the film but it does not bind either by non-specific interactions.
Figure 2. 2. Evolution of the normalized frequency shift measured by QCM at 15 MHz (third harmonic) during the buildup of PEI(PSS/PAH)3/PAA/PAH/PAA as a function of time. a) Film put in contact
with streptavidin. b) Film put in contact with serum.
In order to obtain films that prevent non specific adsorption of proteins and induce specific adsorption of streptavidin, we have adsorbed the modified PAA’s on top of the PEM films. Depending on the nature of the top layer, two strategies were developed:
1) In the first strategy, the top layer is PAA(EO)nBiotin modified with different spacer
45
2) In the second strategy, the top layer is PAA modified both with PC and EOnBiotin with
respective GR of 25 % and 1 %, and with different (EO)n spacer lengths. 2.3.2 PAA(EO)nBiotin ending films
Strategy 1.
Figure 2. 3. Scheme of functionalized PEM films with PAA(EO)nBiotin, n = 0, 3, 9 and 18.
We built the following films : PEI(PSS/PAH)3PAA/PAH/PAA(EO)nBiotin with n = 0, 3, 9
and 18 and with GR = 1, 5, 10 and 25 %. The adsorption of streptavidin (0.1 mg/mL) was monitored in situ by QCM. The mass adsorbed of streptavidin was calculated for QCM data using the Sauerbrey relation. The values are reported in Table 2.2 and plotted in Figure 2.4.
Figure 2. 4. Mass adsorbed of Streptavidin on PEI(PSS/PAH)3/PAA/PAH/PAA(EO)nBiotin with GR =
1, 5, 10 and 25 % and n= 0, 3, 9 and 18, calculated from QCM data using the Sauerbrey relation. 0 1000 2000 3000 4000 PAA-B PAA(EO)3B PAA(EO)9B PAA(EO)18B 1 5 10 25 M a ss a d so rb e d o f St re p ta v id in (n g /c m 2) Grafting ratio %
46
With PAA-Biotin (w/o spacer), streptavidin adsorption is similar for a GR of 1 and 5 % and decreases for a GR of 10 %. The thickness of adsorbed PAA-Biotin increases with GR. The decrease of the adsorbed streptavidin mass is probably due to a non accessibility to biotin due the lack of spacer between PAA and biotin.
With PAA(EO)3Biotin, the mass of adsorbed streptavidin increases when GR increases
from 1 to 10 % and has the same value at 25 than 10 %. With PAA(EO)9Biotin and
PAA(EO)18Biotin, the mass of adsorbed streptavidin increases with GR. The highest adsorbed
amount of streptavidin is found for PAA(EO)9Biotin with a GR of 25 %.
To test the anti-fouling property of the films, they were brought in contact with serum at 4.4 mg/mL. The results are illustrated in Figure 2.5 and summarized in Table 2.2.
Figure 2. 5. Mass adsorbed of Serum on PEI(PSS/PAH)3/PAA/PAH/PAA(EO)nBiotin with GR = 1, 5,
10 and 25 % and n= 0, 3, 9 and 18, calculated from QCM data using the Sauerbrey relation.
The adsorption of serum on PAA-Biotin increases when GR increases. On PAA(EO)3Biotin it is independent for the GR. On PAA(EO)9Biotin and PAA(EO)18Biotin it
decreases drastically when GR increases. At GR = 25 %, the serum adsorption was equal to the detection limit of the QCM.
0 200 400 600 800 1000 1200 1 5 10 25 M a ss a d so rb e d o f Se ru m ( n g /c m 2) Grafting ratio % PAA-B PAA(EO)3B PAA(EO)9B PAA(EO)18B
47
Table 2. 2. Mass adsorbed of Streptavidn and Serum on PEI(PSS/PAH)3/PAA/PAH/PAA(EO)nBiotin
films, calculated from QCM data by applying the Sauerbrey relation.
Nature of the top layer deposited on PEI/(PSS/PAH)3PAA/PAH GR in % Streptavidin adsorption (ng/cm2) Serum adsorption (ng/cm2) PAA - <1±1 1252±53 PAA-B 1 2165±218.5 702±59.9 5 1990±87.3 765±41.8 10 1241±43.2 920±98.5 PAA(EO)3Biotin 1 947±143.2 564±55.6 5 1760±87.5 976±49.7 10 2305±371.3 943±148.9 25 2407±259.6 732±41.7 PAA(EO)9Biotin 1 714±125.2 457±145.2 5 1794±59 258±140.5 10 1483±605.6 40±39 25 3192±295.7 <1±1 PAA(EO)18Biotin 1 281±3.5 723±69.2 5 1223±51.3 51±88.9 10 1249±101.4 49±85.4 25 2419±80.1 <1±1
At 25 % in GR, the amount of adsorbed streptavidin is higher on PAA(EO)9Biotin (3192
ng/cm2) than on PAA(EO)18Biotin (2419 ng/cm2).(Table 2.2).
Biotin groups of PAA(EO)18Biotin are probably not all accessible on surface of the film.
This could be explained by the flexibility of the ethylene oxide groups that can fold down.87 In the case of PAA(EO)9Biotin, the protein resistance is attributed to the formation of
brush layers on the surface that act as a steric barrier. While for PAA(EO)18Biotin, the
conformation is more like mushroom like form inhibiting the non specific adsorption of proteins. In the same time the biotin moieties are less exposed towards the supernatant and less accessible to streptavidin.
48
2.3.3 PAA-PC-(EO)nBiotin ending films
As described by Reisch et al.68, 69, PAA-PC with GR = 25 % has anti-fouling properties. Therfore, we have used the same polymers, with the same GR in PC. But in addition we have further functionalized them with (EO)nBiotin with GR = 1 % and we have varied the length of
(EO)n.
Strategy 2.
Figure 2. 6. Scheme of functionalized PEM films with PAA-PC-(EO)nBiotin, n = 0, 3, 9 and 18.
After the adsorption of PAA-PC-(EO)nBiotin on the top of the films, the completed films
were incubated with serum at 4.4 mg/mL or streptavidin solution at 0.1 mg/mL. The results are displayed in Figure 2.7 and Table 2.3.
Table 2. 3.Mass adsorbed of Streptavidin and Serum on PEI(PSS/PAH)3
/PAA/PAH/PAA-PC-(EO)nBiotin films, calculated from QCM data by applying the Sauerbrey relation
Nature of the top layer deposited on PEI/(PSS/PAH)3PAA/PAH GR % Streptavidin adsorption (ng/cm2) Serum adsorption (ng/cm2) PAA-PC/Biotin 25/1 342±8.1 <1 PAA-PC/(EO)3Biotin 25/1 449±64.6 <1 PAA-PC/(EO)9Biotin 25/1 588±166.1 <1 PAA-PC/(EO)18Biotin 25/1 445±69.2 <1
49
Figure 2. 7. Mass adsorbed of Streptavidin on PEI(PSS/PAH)3/PAA/PAH/PAA-PC-(EO)nBiotin
calculated from QCM data using the Sauerbrey relation.
The amount of serum adsorbed for the different films was of the order of the detection limit of the QCM (Table 2.3), which is in agreement with the results of PAA-PC with GR of 25 % reported by Reisch et al.69
Figure 2.7 displays the mass adsorbed of streptavidin obtained for PAA-PC-(EO)nBiotin
ending film, with different lengths of oligo(ethylene oxide) (EO)n (n = 0, 3, 9 and 18). The
amount of bound streptavidin increases with the spacer length. Streptavidin adsorbs less on PAA-PC-(EO)18Biotin than on PAA-PC-(EO)9Biotin. This feature can be explained by the
decrease of accessibility to the biotins at the surface due to the fold down of the EO chains. The optimal adsorption is thus found for PAA-PC-(EO)9Biotin.
These surfaces specifically adsorbs streptavidin with higher efficiency for PAA-PC-(EO)9Biotin and at the same time prevent non-specific adsorption. This property can be
explained by the presence of PC and EO groups that are well known to have anti-fouling properties. EO displayed anti-fouling properties as well as PC groups.69 Non specific adsorption of protein decreases with the increase in length of EO and GR.
0 100 200 300 400 500 600 700
(EO)0 (EO)3 (EO)9 (EO)18
M a ss a d so rb e d o f St re p ta vi d in (n g /c m 2)
50 2.4 Conclusion
In this study, two strategies were applied to obtain a specific adsorption of streptavidin and to prevent the non specific adsorption of serum proteins. Films with PAA-PC-(EO)nBiotin and
PAA (EO)nBiotin as last layers specifically adsorb streptavidin. The amount of protein adsorbed
depends on the EO length and the modification degree (GR) in linker. The highest ratio of streptavidin adsorption over non specific adsorption is reached for (EO)9.
The non-specific adsorption of serum proteins on PAA-PC-(EO)nBiotin is totally precluded
for any (EO)n linker length. The polyelectrolytes PAA(EO)nBiotin resist to non specific
adsorption when the length is 9 or 18 and the GR is 25 %. In both strategies, modified PAA with (EO)9 as linker shows the best results, i.e the highest specific adsorption of streptavidin and the
smallest non-specific adsorption.
When comparing both strategies, the films with PAA(EO)9Biotin are able to bound higher
amounts of streptavidin than PAA-PC-(EO)9Biotin. In addition, PAA(EO)9Biotin with GR of 25
% is resistant to non specific protein adsorption. This behavior can be explained by the structure of the modified polyelectrolyte on the film surface. The EO groups are strongly hydrated and form a steric barrier as loop-rich layer being most efficient against protein adsorption.
The designed surfaces are a powerful technique to obtain specific adsorption of proteins and to be good candidates for the development of biosensors.
Indeed by using both strategies it is possible to immobilize a biotinylated antibody that can recognize its antigen. (Figure 2.8)
52
3
CHAPTER 3:
Cyto-mechanoresponsive
Polyelectrolyte Multilayer Films
53
3.1 Article 1: Cyto-mechanoresponsive Polyelectrolyte Multilayer Films
In this chapter, we have studied films that are able to become cell adherent or to induce specific adsorption of protein by the application of a mechanical stimulus. The design of surfaces that respond under mechanical stimulus has become very attractive field. Only few mechanosensitive surfaces have been described in the literature. One of the first studies was the development of surfaces that present mechanochromic behavior,98,99 which are able to change their color upon stretching.
In nature, the proteins change their conformation under mechanical forces, exhibiting cryptic sites, e.g. fibronectin are involved in this process where the cryptic sites are exposed under stretching.54-56 This concept have inspired the development of surfaces containing active compounds which are inaccessible in non stretched state and become accessible when the surface is stretched.56
Summary
In this section, we have designed surfaces that exhibit different features, cell adhesion and specific adsorption of proteins when they are stretched. To reach this goal, silicon substrates were used as substrate of PEM films.
In the case of cell adhesion property, PAA was modified by arginine-glycine-aspartic acid (RGD) sequence to obtain PAA-RGD. In order to induce the cell adhesion under stretching, the silicon surfaces were functionalized by PEM films composed by PAA-RGD and PAA-PC. The buildup consist on PEI(PSS/PAH)3 as a precursor film, followed by PAA-RGD embedded by 2
to 6 bilayers of (PAH/PAA-PC) to obtain PEI(PSS/PAH)3/PAA-RGD/(PAH/PAA-PC)n with n =
2 to 6. It was found that in non stretched state (PAH/PAA-PC)2 film adsorbed onto
PEI(PSS/PAH)3/PAA-RGD is needed to render the surface totally resistant to cell adhesion.
Furthermore by stretching the functionalized silicon up to 150 % of its original length, cell adhesion is observed.
In the second part of this section, the specific adsorption of proteins was also studied as for cell adhesion, but in this case using modified PAA with biotin moieties. The strong interaction between streptavidin-biotin was used as model system. First of all the film buildup was studied
54
by QCM, with PEI(PSS/PAH)3/PAA-(EO)nBiotin/PAH/PAA-PC (n = 0, 3 and 9). The results
demonstrated that one layer of PAA-PC was enough to prevent the adsorption of streptavidin in (EO)n (n = 0 and 3). Then the studies with these moieties were carried on silicon substrates to
determine their behavior under stretching. Using fluorescein labeled streptavidin (StrepFITC) and
fluorescence microscopy in non stretched state, the findings were in good agreement with the results obtained by QCM. When the film is stretched, a strong streptavidin adsorption was observed, increasing with the degree of stretching. The reversibility of the films was also studied. When the film is stretched and put back at its non-stretched state, the presence of streptavidin reveals fluorescence intensity. The same behavior is obtained for adhesion of cells on RGD functionalized films. This system seems not reversible.
